This thesis studies inventory and production management and performance eval-uation in assemble-to-order (ATO) systems. We propose profit-maximization models in various settings that are motivated from industrial practices. We pro-vide efficient solution procedure to solve the proposed problems. We also offer managerial insights that are potentially useful for managing ATO systems. In addition, we propose efficient approximation methods to evaluate performance measures in a capacitated continuous time ATO system. In the following we briefly introduce three major models studied in this thesis....[ Read more ]

This thesis studies inventory and production management and performance eval-uation in assemble-to-order (ATO) systems. We propose profit-maximization models in various settings that are motivated from industrial practices. We pro-vide efficient solution procedure to solve the proposed problems. We also offer managerial insights that are potentially useful for managing ATO systems. In addition, we propose efficient approximation methods to evaluate performance measures in a capacitated continuous time ATO system. In the following we briefly introduce three major models studied in this thesis.

In the first problem we consider an inventory and production planning problem for a contract manufacturer who anticipates an order of a single product with uncertain quantity. To meet the challenges of long component procurement lead times and limited assembly capacity, which may render production time insuf-ficient to assemble total order quantity, the manufacturer may need to procure components or even assemble some quantities of the final product before receiv-ing the confirmation of the actual order quantity. To maximize the total expected profit, we establish structural properties of optimal solutions and develop effi-cient solution procedures. We also provide sensitivity analysis of the optimal decisions and some managerial insights.

In the second problem, we consider a component acquisition problem for a con-tract manufacturer who faces a one-time stochastic demand of a single product consisting of multiple components. Components can be ordered early at normal prices before the demand is revealed, or they can be replenished later at higher prices due to expediting. In addition, the customer pays a price for the final product that is decreasing in the delivery lead time to discourage late delivery. The manufacturer needs to make trade-offs between stocking too many compo-nents at normal prices and having to expedite the missing components at higher prices and losing revenue due to delays caused by component replenishment lead times. We propose a cost-minimization model to address the component stocking decisions. We develop structural properties and propose an efficient algorithm that solves the problem. We also analyze the optimal decisions’ sensitivity to changes of lead times and component costs under expediting.

In the last problem, we consider performance evaluation for a multi-component multi-product assemble-to-order system. Each component supply facility has exponential processing time and infinite buffer and each component is managed independently using base-stock policy. Since exact analysis of the system is difficult, we propose two approximation methods to derive approximations for the performance measures such as fill rate, average waiting time, and average number of backorders. We show that the two approximation methods provide upper and lower bounds of the real system. Computational studies demonstrate that our methods provide very accurate approximations.